Reconstruction of the Three-Dimensional Structure of the Human Brain Using Magnetic Resonance Imaging and Deep Learning

• A deep learning framework combining GAN and spatial attention achieves superior 3D brain reconstruction from low-resolution MRI, with PSNR of 32.5 dB and SSIM of 0.94. • The method preserves anatomical details critical for accurate cortical thickness measurement and lesion detection, outperforming conventional super-resolution approaches. • The reconstructed 3D models show high fidelity, enabling reliable clinical applications such as surgical planning and neuroimaging diagnostics. • The framework demonstrates robustness across diverse MRI datasets, suggesting broad applicability in medical imaging.